This grant proposal is about the development of the palate, with special attention to the role of the medial edge epithelium (MEE) and cranial neural crest mesenchyme. Mesenchymal cells arise by epithelial-mesenchymal transformation (BMT). They have the ability to invade collagen gels and other extracellular matrix (ECM), whereas epithelia sit on top of ECM and exhibit apical-basal polarity. Mesenchymal cells migrate through the matrix with front end/back end polarity. They differentiate into the cells that inhabit the ECM interior: fibroblasts, bone, melanocytes, and neurones. We will consider the following aspects of the origin and differentiation of the craniofacial mesenchymes that contribute to palatal morphogenesis. I. Mesenchymal Cell Lineages in the Developing Palate. Using recombinant retroviruses to label specific groups of cells in the embryo, we expect to confirm and extend our earlier study that MEE-derived mesenchyme forms fibroblasts but not bone, and to prove for the first time that head neural crest in the rodent consists of two physically separate cell populations, one of which forms connective tissue and the other, melanocytes and neurones. The significance of this study to students of palate morphogenesis is that it establishes a multiple origin of palate mesenchyme and calls attention to the fact that palate MEE does not die as previous authors thought. It casts a new light on roles the MEE might play in cleft palate. II. Effect of Bone Inducing Molecules on the Differentiation of Mesenchymal Cell Lineages. We will attempt to drive the cells above that do not normally form cartilage/bone to do so by administering BMP to them in vitro. It will be particularly interesting to see if mesenchyme formed from palate epithelium and neurogenic crest cells will undergo chondrogenesis under the influence of this powerful inducing agent. The results could have important impact on our ideas about the ability of "determined" cells (e.g., neurogenic crest) to reawaken apparently closed pathways of differentiation. III. Induction and Inhibition of Epithelial Fusion and EMT in Nonfusing and Fusing Palates. Chicken palates do not normally fuse probably because the MEE stratifies and periderm does not slough to expose the basal epithelia to each other. We will surgically remove the periderm and place chicken palatal shelves in close contact to try to induce formation of a midline seam. TGFbeta3 will be added to promote EMT from this seam and the fate of mesenchymal cells followed. Antibodies and antisense oligonucleotides to TGFbeta1-3 will be added to try to inhibit EMT from MEE of fused rodent palates. The recognition our recent work has given to the fact that MEE transforms to mesenchyme establishes a whole new way of looking at the possible role of TGFbeta's and sloughing of the outer layer of the MEE in palatogenesis. The immediate health relevance is to understanding normal development, but the work could give new insights into causes of congenital anomalies, especially cleft palate.